The benefits of characterizing an acoustic field by measuring particle motion, instead of local changes in ambient pressure, have long been known. Measurement of the particle motion in terms of displacement, velocity or acceleration reveal information on the field direction as well as amplitude. Simultaneous measurement of the local pressure, p, and particle velocity, {dot over (u)}, in a plane wave also yields the acoustic intensity through the relationship, I=<real(p)real({dot over (u)})>r, where <>r denotes time average. Here, u, is particle displacement and u denotes differentiation with respect to time. When it is of interest to detect and locate sources of acoustic radiation through the use of coherently beamformed arrays, particle motion sensors provide further benefits due to their inherent directionality. An array of three-axis directional sensors achieves equal array gain to an array of scalar hydrophones with twice the length. Furthermore, the left-right ambiguity arising from the symmetry of scalar sensor line arrays is overcome with arrays of directional sensors.
Most sensors designed to respond to acoustic particle motion are based on some form of simple harmonic mechanical oscillator that is driven by the acoustically induced motion. The operating bandwidth of the sensor is determined by the fundamental resonant frequency of the oscillator. Assuming the sensor responds linearly to the relative displacement between the inertial mass and its case, then operation below the fundamental resonance will result in a response to acceleration independent of frequency (i.e. an accelerometer). Operation above the fundamental resonance results in a response to displacement independent of frequency. One such sensor, known as the moving coil geophone, that operates above the fundamental resonance responds to the rate of change of the sensor easing and thus responds to velocity independent of frequency.
For measuring low frequency (i.e. <10 kHz) acoustic fields in the ocean, a frequency independent or smooth response to particle velocity is highly desirable, in a planar wave field, the acoustic impedance relating the ratio of the pressure to particle velocity is given by the product of fluid density and sound speed, ρflcfl. Thus, the ambient velocity noise field is proportional to the pressure field. This is beneficial since the spectral density of the ambient acoustic pressure noise exhibits an approximately, 1/f dependence, which is a similar dependence to the internal electronic noise spectrum in many sensors, particularly at low frequencies. Consequently, when the limiting noise source is ambient acoustic, the acoustic resolution is not diminished for deceasing frequency with a true velocity sensor. The interest in developing low frequency vector sensors is apparent from the number of reported devices. For example, devices based on moving coil, piezoelectric, resistive heating, and magnetorestrictive mechanisms have been demonstrated as well as several fiber-optic devices based on pressure gradient hydrophones. However, achieving high acoustic sensitivity over a large bandwidth from a small sensor remains a challenge.